 Indra is a system designed to extract meaningful information from scientific publications and combine it with existing knowledge bases such as protein-protein interaction databases and pathway databases. It does this by leveraging multiple natural language processing systems to identify overlapping information between different sources improving the accuracy of the extracted data. This allows for the creation of more comprehensive and reliable protein-protein interaction databases which can then be used to build a cancer dependency map that explains how cancer cells interact with their environment. This article was authored by John A. Bachman, Benjamin M. Giore, and Peter K. Sorge.